Description of Acoustic and Articulatory Parameters of Vowels in Mapudungun Speakers. Pilot Study - SciELO
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Int. J. Odontostomat.,
14(2):205-212, 2020.
Description of Acoustic and Articulatory Parameters
of Vowels in Mapudungun Speakers. Pilot Study
Descripción de Parámetros Acusticos y Articulatorios de las
Vocales en Hablantes de Mapudungun. Estudio Piloto
Giannina Álvarez1; Magaly Ruiz2; Alain Arias1,3,4; María Florencia Lezcano1,3 & Ramón Fuentes1,3
Álvarez, G.; Ruiz, M.; Arias, A.; Lezcano, M. F. & Fuentes, R. Description of acoustic and articulatory parame-
ters of vowels in Mapudungun speakers. Pilot study. Int. J. Odontostomat., 14(2):205-212, 2020.
ABSTRACT: Mapudungun is a language used by Mapuche people in some regions of Chile and Argentina. The
aim of this study was to describe the vowel phonemes with regard to the articulatory parameters (position of the tongue with
respect to the palate and jaw opening) and acoustic parameters (f0, F1, F2 and F3) in Mapudungun speakers in the Region
of La Araucanía. The vocalic phonemes of Mapudungun are six, where the first five are similar to those used in Spanish
(/a e i o u/), to which is added a sixth vowel (/ɨ/) with its vocalic allophones (/ɨ/) and [Ә]. Three Mapudungun speakers were
evaluated. The tongue movements were collected by Electromagnetic Articulography 3D and the data were processed with
MATLAB and PRAAT software. It was possible to describe the trajectory of each third of the tongue during the production
of the vowels. It was observed that the sixth vowel /Ә/ had minimal jaw opening during its pronunciation. In addition, the
characteristic of /Ә/ as an unrounded mid-central vowel was corroborated. In this study, the tongue of mapudungun speakers
was in a more posterior position than the found in other studies.
KEY WORDS: acoustic phonetics; articulatory phonetics; mapudungun; electromagnetic articulography;
PRAAT
INTRODUCTION
Mapudungun is a language used by the Mapu- rounded or not rounded in their labialization (Ladefo-
che in some regions of Chile and Argentina (Echeverría ged, 2001). From this perspective, the phoneme /a/
& Contreras, 2005; Zúñiga, 2006). For the phonetic corresponds to a low central unrounded vowel, the
study of this language, as well as for the articulatory phoneme /e/ is defined as a back mid unrounded vowel,
and acoustic analysis of the voice, the phonemes of /i/ is a high back unrounded vowel, /o/ is a mid-back
Mapudungun can be classified as vowels, semi-vowels rounded vowel and /u/ is formed as a high back rounded
and consonants (Zúñiga). In the case of the vowels, vowel. In the case of /ɨ/ for its allophones /ɨ/ and /Ә/,
there are six vocalic phonemes where the first five the first is a high back rounded vowel, while the second
are similar to those used in Spanish (/a e i o u/) with is a mid-central unrounded vowel (Sánchez Perez &
identical allophones, to which is added a sixth vowel Salamanca Gutierrez, 2015; Echeverría & Contreras;
(/ɨ/) with its vocalic allophones (/ɨ/) and [Ә] (Echeverría, Soto-Barba et al., 2016).
1964; Echeverría & Contreras; Sadowsky et al., 2013).
In acoustic phonetics, the vowels can be
Each of these phonemes presents distinctive analyzed on the basis of their fundamental frequency,
articulatory features in terms of their vertical dimension intensity, formants and measures of disturbance like
(high, mid and low vowels), their horizontal dimension jitter and shimmer, among others (Franca, 2012). The
(front, central and back vowels) and whether they are acoustic vowel space can be determined by the area
1
Research Centre for Dental Sciences CICO, Dental School, Universidad de La Frontera, Temuco, Chile.
2
Department of Language, Literature and Communication, School of Education, Social Sciences and Humanities, Universidad de La
Frontera, Temuco, Chile
3
Department of Integral Adults Dentistry, Dental School, Universidad de La Frontera, Temuco, Chile
4
Research Group of Health Sciences, Universidad Adventista de Chile, Chillán, Chile.
This research is part of the project DFP18-0037
205Álvarez, G.; Ruiz, M.; Arias, A.; Lezcano, M. F. & Fuentes, R. Description of acoustic and articulatory parameters of vowels in Mapudungun speakers. Pilot study. Int. J.
Odontostomat., 14(2):205-212, 2020.
defined by F2 and F1 (Weirich & Simpson, 2013) and oral cavity: skin point of the right (1) and left mastoid
is used to represent the relative positions of the ton- (2), glabella (3); those on the back of the tongue in
gue, in both its height and front-back advancement its anterior (4), middle (5) and posterior third (6) were
during vowel production (Sun et al., 2013; Wang et al., positioned according to previous protocols (Chen et
2013). Although the phonetic and acoustic analysis of al., 2015; Álvarez et al., 2019), mandibular inter-incisor
Mapudungun has made progress, the articulatory cha- midline (7) and an extra sensor attached to a wooden
racteristics have scarcely been explored. Therefore, the rod to record the midsagittal profile of the palate (8)
aim of this exploratory study was to describe the vowel (Toutios et al., 2011; Álvarez et al.,). The three first
phonemes with regard to the articulatory parameters sensors were used as a benchmark to standardize
(position of the tongue with respect to the palate and the position data for the remaining sensors (4 to 7)
jaw opening) and acoustic parameters (fundamental and thus eliminate the natural movement of the head
frequency and first three formants) in Mapudungun from the records.
speakers in the Region of La Araucanía.
Once the sensors had been placed, and to be-
gin the vowel recordings, the participants were seated
MATERIAL AND METHOD under the transmitter coils of the articulograph in an
upright, comfortable position facing front.
Participants: In this preliminary study, acoustic data Audio recording: At the same time as the articulatory
and tongue movement (articulatory) data were recor- movement was recorded, a voice recording (.wav for-
ded in three Mapudungun speakers (2 men and 1 mat) was made using the system incorporated into the
woman), with an average age of 26.3 ± 4 years (Table articulograph, which includes an ALESIS i02 portable
I). All the participants signed the respective informed audio interface with a t-bone EM 9600 unidirectional
consent to take part in the study, which was approved microphone (frequency range 60 Hz – 18 KHz) placed
by the Scientific Ethics Committee of the Universidad 30 cm from the participant’s mouth.
de La Frontera (Protocol Nº 125/18). The speakers had
full dentition with permanent teeth (up to the first molar), Recording procedure: Prior to the recording session
occlusal stability and no respiratory problems or mor- with each participant, each was asked to remove any
phofunctional alterations of speech, voice or hearing. metallic elements and electronic devices they had on
them (to avoid interference with the equipment). Each
The participants are speakers from the provin- participant was instructed on the activities to perform
ce of Cautin, Region the La Araucanía, bilingual, whose during the measurement, and training was done to
mother tongue is Spanish (L1) and who received formal become familiar with the presence of the sensors on
or informal instruction in Mapudungun during their child- the tongue, (first without the sensors and then with
hood. All the speakers in this study use Mapudungun the sensors positioned on the tongue) consisting of
(L2) on a regular basis to communicate in their family breathing exercises, swallowing saliva, articulation of
unit and are recognized in their communities for their the vowel phonemes and four specific words (atraso -
proficiency in and teaching of this language. angosto - pantruca - angina). These training words were
selected for having: a) three syllables, b) accent on the
Electromagnetic articulography (EMA 3D): The penultimate syllable (paroxytone words) and c) stressed
acoustic data and tongue movement, its relation with syllables joined as CCV (consonant, consonant, vowel)
the palate and jaw opening, were simultaneously re- or CVC (consonant, vowel, consonant).
corded during the articulation of the vowels in Spanish
and Mapudungun: /i/, /e/, /a/, /o/, /u/, /Ә/. The 3D AG501 Thus the following records were made:
electromagnetic articulograph (Carstens Medizinele-
ktronik, Bovenden, Germany) was used, which has i) Record of the midsagittal profile of the palate, po-
suitable spatial resolution (margin of error 0.3 mm) and sitioning the sensor of the wooden rod at the base of
temporal resolution (sampling frequency to 250 Hertz) the participant’s uvula to then advance forward from
to record and analyze articulatory movements (Steele the soft and hard palate, via the midline, until arriving
& Van Lieshout, 2004; Fuentes et al., 2017). Eight at the incisive papilla.
sensors from the articulograph were attached with a
tissue adhesive (Epiglu®, Meyer-Haake, Germany) to ii) Record of the maximum intercuspation position (MIP)
different specific points on each participant’s head and to define a starting point with dental contact between
206Álvarez, G.; Ruiz, M.; Arias, A.; Lezcano, M. F. & Fuentes, R. Description of acoustic and articulatory parameters of vowels in Mapudungun speakers. Pilot study. Int. J.
Odontostomat., 14(2):205-212, 2020.
the mandible and the maxilla. Thus it was possible to N1, and the lowest observed value was 636.4 mm2 for
determine the width of the jaw opening when pronoun- speaker N3, both men. In the same table (Table I), the
cing each vowel. residential area (urban or rural) is indicated.
iii) Record of the vowel phonemes (/a/, /e/, /i/, /o/, /u/, Table II provides the values (in mm) of the maxi-
/Ә/). For this, an operator instructed the participant to mum jaw opening for each of the vowels in the three
pronounce each vowel one by one and maintain it for participants. This value was calculated in relation to the
at least 2 seconds with a tone of voice they usually MIP record, with the highest value being for the vowel
use. There were three repetitions per phoneme and /a/ of participant N1 (11.09 mm) and the lowest value for
the most intelligible one was selected for the analysis. the vowel /Ә/ of participant N2 (1.86 mm). Additionally,
the distances or maximum and minimum separation (in
Data processing: The MATLAB® (MathWorks Inc., mm) of each lingual third (anterior, middle and posterior)
USA) software was used to process the position data relative to the palate during the vocalization sequence
(articulatory movements) obtained with the articulo- of the three participants are shown. The greatest dis-
graph. With this software it was possible to generate tance for the anterior third of the tongue was during the
specific calculation routines or processing scripts and articulation of the vowel /o/ and it was 32.11 mm, for the
thus to collect quantitative data on the movement of middle third of the tongue language it was during the
the tongue in its three thirds (anterior, middle and articulation of the vowel /o/ and it was 32.56 mm, while
posterior) and of the jaw in the sagittal plane (superior for the posterior third of the tongue it was during the
- inferior). Also, the shape and dimension of the pala- articulation of the vowel /u/ and it was 28.66 mm, all of
te was obtained to calculate the area (in mm2) of the them in participant N1. On the other hand, the smallest
sub-palatal space in the sagittal plane. For this study it distance for the anterior third was during the articulation
was defined as the area between the sub-palatal space of the vowel /u/ and it was 0.49 mm, for the middle third
and the straight line that joins the first and last point it was during the articulation of the vowel /i/ and it was
of the palate delimitation record (between the base of 0.00 mm (tongue in direct contact with the palate), while
the uvula and incisive papilla). It was also possible to for the posterior third it was during the articulation of
create trajectory diagrams of the lingual thirds and jaw all the vowels with the exception of the vowel /a/ and it
relative to the palate with the software. was 0.01 mm in all those cases. All these values were
in participant N2.
The PRAAT free software was used to process
the voice records (Boersma and Weenink, Amsterdam, Figure 1 presents the means of the jaw dis-
2018). Thus, an acoustic analysis was performed of the placements (in mm) relative to starting point or MIP,
fundamental frequency, formants and acoustic vowel where a trend is observed of greater jaw opening for
space for each vowel pronounced. the vowel /a/ and smaller jaw opening for the vowel /Ә.
The same chart also shows the mean of the minimum
and maximum distances of the lingual thirds relative
RESULTS to the palate, where it is possible to note a tendency
towards greater descent of the tongue for the anterior
third (with average maximum distances greater than
The size of the sub-palatal space is itemized in the other thirds) and a smaller distance or separation
Table I, where, by calculating the area in the midsagittal relative to the palate (greater contact with the palate)
plane, a reference of the height of the palate with respect for the posterior third.
to its front and back edges was obtained. On this point,
the highest observed value was 939.5 mm2 for speaker Figure 2 provides an example of the trajectory
Table I. Characteristics of the study participants.
Sagittal area of
Age Area of
Participant Sex L1 L2 the sub-palatal
(years) residence
space (mm)2
N1 Male 24 Spanish Mapudungun Rural 939.5
N2 Female 31 Spanish Mapudungun Urban 815.6
N3 Male 24 Spanish Mapudungun Rural 636.4
207Álvarez, G.; Ruiz, M.; Arias, A.; Lezcano, M. F. & Fuentes, R. Description of acoustic and articulatory parameters of vowels in Mapudungun speakers. Pilot study. Int. J.
Odontostomat., 14(2):205-212, 2020.
Table II. Width of the jaw opening (mm) relative to the MIP and minimum and maximum distances (separation) of the
lingual thirds (anterior, middle and posterior) in millimeters relative to the palate during the articulation of each vowel.
Tongue Tongue Tongue Tongue Tongue Tongue
Jaw anterior anterior middle middle posterior posterior
Vowel Participant
opening third min third max third min third max third min third max
(mm) (mm) (mm) (mm) (mm) (mm)
N1 11.09 5.91 26.28 9.18 27.45 6.04 16.67
/a/ N2 6.90 8.40 20.75 4.63 11.17 2.52 5.43
N3 8.08 3.66 17.39 0.55 13.26 2.31 12.20
N1 7.17 6.64 14.81 11.65 14.14 6.15 8.80
/e/ N2 4.64 7.65 12.07 0.27 3.48 0.01 4.84
N3 9.17 2.63 15.80 0.97 6.24 1.92 11.64
N1 4.38 8.07 12.08 8.09 14.06 3.63 10.33
/i/ N2 5.53 4.87 11.57 0.00 2.17 0.02 5.93
N3 6.55 2.37 12.61 1.05 5.46 3.04 13.74
N1 7.27 8.94 32.11 18.44 32.56 12.60 19.58
/o/ N2 6.27 13.10 21.58 6.80 12.42 0.01 3.83
N3 5.13 2.47 20.11 3.51 12.20 0.42 7.50
N1 3.22 4.89 19.32 14.86 25.46 18.86 28.66
/u/ N2 5.77 0.49 19.52 3.67 7.70 0.01 3.43
N3 5.58 1.24 23.18 0.08 11.71 0.02 7.84
N1 4.94 6.72 12.72 14.78 24.78 5.60 11.68
/Ә/ N2 1.86 9.97 14.04 2.90 6.93 0.01 5.28
N3 4.16 1.35 24.42 0.15 18.19 0.85 8.45
Fig. 1. Diagram with the displacement (in mm) of the open jaw and distances to the palate (mm) of the tongue thirds during
the vocalization of /i/, /e/, /a/, /o/, /u/, /Ә/.
208Álvarez, G.; Ruiz, M.; Arias, A.; Lezcano, M. F. & Fuentes, R. Description of acoustic and articulatory parameters of vowels in Mapudungun speakers. Pilot study. Int. J.
Odontostomat., 14(2):205-212, 2020.
in the sagittal plane of the sensor placed on the jaw the vocalization of /a/, /e/, /i/, /o/, /u/, /Ә/. Note that for
and of the sensors placed on the anterior, middle and the case of /a/, there is a drop in the central position of
posterior thirds of the tongue in participant N1 during the lingual thirds, accompanied by the jaw opening. In the
Fig. 2. Position diagram with the trajectory in the sagittal plane (superior-inferior and posterior-anterior axis) of the
sensors located in the anterior, middle, and posterior thirds of the tongue (red lines), the sensor of the jaw (blue
line) and the sagittal profile of the palate (dashed black line) of one of the participants (N1) during the vocalization
of /a/, /e/, /i/, /o/, /u/, /Ә/. The circle indicates the initial position of the sensor during the recording.
209Álvarez, G.; Ruiz, M.; Arias, A.; Lezcano, M. F. & Fuentes, R. Description of acoustic and articulatory parameters of vowels in Mapudungun speakers. Pilot study. Int. J.
Odontostomat., 14(2):205-212, 2020.
case of /e/, there is a minimal drop of the anterior third Hz during the articulation of the /u/ vowel by participant
accompanied by a similar drop of the jaw, a backward N2, whereas the smallest value found was 105.5 Hz
advancement of the middle third and a slight rise of during the articulation of the /a/ vowel by participant N3.
the posterior third. For /i/, close contact of the middle For F1, which indicates the height of the tongue, the
and posterior thirds with the palate is observed, with greatest value found was 728 Hz for the articulation of
a marked upward and forward displacement and a the /a/ vowel by participant N2, whereas the smallest
slight drop of the anterior third similar to the downward value found was 289.2 Hz for the articulation of the
displacement of the jaw. In this case, contact with the /i/ vowel by participant N3. With respect to F2, which
palate is noted by the demarcation and limitation in the indicates the anterior-posterior position of the tongue,
movement that the sensors present, but graphically this the greatest value was 2532 Hz for the articulation of
contact is not faithfully observed due to the size of the the /i/ vowel by participant N2, whereas the lowest
sensor (3 mm), whose movement recording is given value was 849 Hz for the articulation of the /u/ vowel
from its center. In the case of /o/ the displacement is by participant N1. Finally, for F3, which informs on the
downward, similar to that of the /a/ vowel but towards roundedness of the lips, the highest value was 3312
a posterior position, whereas the jaw drop is slight, not Hz for the articulation of the /a/ vowel by participant
accompanying the greatest width that the sensors show N2, whereas the smallest value was 2556 Hz for the
of the anterior and middle thirds. In the case of /u/, a articulation of the /e/ and /Ә/ vowels by participant N3.
similar drop and rise of the anterior lingual third and the
jaw is observed, a noticeable downward and upward Figure 3 presents the diagram of the acoustic
displacement of the middle third and an upward and vowel space obtained from the values of F1 and F2 of
backward displacement of the posterior third. Finally, the vowels /a, e, i, o, u, Ә / for the three Mapudungun
for /Ә/ a slight drop in the anterior third, posterior third speakers. In the scheme obtained, there is a certain
and the jaw is observed, and a large drop in the middle variability among the speakers, with the female speaker
third. (N2) being the one who moves away at some points
from the other speakers. With her, there was a lower
Regarding the data collected of the base fre- tongue height and pharyngeal narrowing like a greater
quencies (f0) and the formants F1, F2 and F3, Table anteriorization of the tongue for all the vowels compa-
III shows the values for each vowel individually per red to the other speakers, which translates into higher
participant. For f0, the greatest value found was 204 frequencies in the emission of the vowels.
Table III. Values in Hertz of the fundamental frequency and the three first formants for each vowel.
Vowel Participant f0 (Hz) F1(Hz) F2(Hz) F3(Hz)
N1 125.3 679.3 1411.0 2654.0
/a/ N2 174.3 728.0 1459.0 3312.0
N3 105.5 654.9 1264.0 2581.0
N1 119.7 459.9 1825.0 2629.0
/e/ N2 184.9 459.0 2337.0 2873.0
N3 112.3 435.0 2093.0 2556.0
N1 121.7 362.4 2191.0 2702.0
/i/ N2 185.1 435.5 2532.0 2971.0
N3 122.0 289.2 2410.0 2849.0
N1 125.8 484.2 874.3 2824.0
/o/ N2 190.0 557.4 1142.0 3214.0
N3 108.4 533.0 947.0 2581.0
N1 127.4 411.1 849.0 2751.0
/u/ N2 204.0 435.0 1094.0 3044.0
N3 120.5 386.0 898.0 2629.0
N1 126.0 435.5 1216.0 2727.0
/Ә/ N2 175.6 435.5 1289.0 2971.0
N3 108.5 459.9 1240.0 2556.0
210Álvarez, G.; Ruiz, M.; Arias, A.; Lezcano, M. F. & Fuentes, R. Description of acoustic and articulatory parameters of vowels in Mapudungun speakers. Pilot study. Int. J.
Odontostomat., 14(2):205-212, 2020.
literature describes the anatomical characteristics of
the vocal tract as being able to influence the production
of sounds (Rendall et al., 2005).
With respect to the acoustic analysis, the values obtai-
ned for the Mapudungun vowels may be comparable
to the data collected by other authors for the vowels in
Spanish and for the sixth Mapudungun vowel. On this
point, we can mention that Sadowsky et al., performed
a standardization of the mean values of the formants
for /Ә/ and its respective allophones, finding for this
vowel average F1 values of 360 Hz and F2 values of
1480 Hz in the Huapi Island sector in the Chile’s Region
of La Araucanía (Sadowsky et al.,; Soto-Barba et al.,).
Fig. 3. Acoustic vowel space, defined by the values of for- Meanwhile, in the study by Soto-Barba et al., for the
mants F1 x F2 of the vowels /i/, /e/, /a/, /o/, /u/, /Ә/ for the analysis of /Ә/ conducted on speakers in the Bío-Bío
three Mapudungun speakers.
and Queuco sectors, Alto Bío-Bío, Chile, they found that
the mean of F1 for each sector was 505 Hz and 475 Hz
DISCUSSION respectively, whereas for F2 it was 1417 Hz and 1350
Hz respectively (Soto-Barba et al.,). In our findings, the
F1 mean /Ә/ was 444 Hz, a value near that reported by
The aim of our study was to describe the Soto-Barba et al., in the Queuco sector (Soto-Barba et
vowel phonemes in terms of articulatory and acoustic al.,). The F2 mean was 1248 Hz in our sample, lower
parameters in Mapudungun speakers. than that described by other authors, which would indi-
cate a less anterior position of the tongue than expected.
Electromagnetic articulography was used to do With respect to the other vowels, a study of the vowels
this because it enables a suitable characterization of in Spanish speakers in Santiago de Chile (Díaz et al.,
the tongue movement and jaw opening during the pro- 2015) indicates higher average fundamental frequency
duction of the vowel sounds. This technique has been values for men and women than those found in this
widely used in phonetics and linguistics (Wang et al., study for all the vowels. In the same study, the F1 values
2009, 2013; Meenakshi et al., 2014; Lee et al., 2017), described for the men were similar or slightly lower than
and in this case it gave us relevant data regarding the those found in our study; however, in the case of the
interaction of the tongue and palate, both from the women, the F1 described here from the female speaker
charted trajectory of the tongue (in its anterior, middle was slightly higher for the extreme vowels /a/, /i/, /u/. In
and posterior thirds) as well as from the quantitative relation to F2, for the vowels /a/, /e/ and /i/, the values
data of the distance of each lingual third to the palate. described in our study were similar or slightly lower than
In the literature it is possible to find different studies those reported by Díaz et al., whereas for /o/ and /u/, the
that consider the position of the palate and the tongue F2 values described in our study are much lower, with
in the production of sound (Ishida et al., 2002; Toutios a mean of 1431 Hz and 1440 Hz respectively for each
et al.; Noiray et al., 2014); however, to the best of our vowel. This indicates that in the Mapudungun speakers
understanding it was not possible to find studies that included in this study, the tongue is in a more posterior
mention an analysis with the minimum and maximum position than expected. In the case of F3, the values of
distances to the palate or to the jaw opening, which the Mapudungun speakers were similar or slightly lower
is why this study contributes new information and a for all the vowels compared to the data found by Díaz et
reliable and replicable methodology for other analyses. al., In the work carried out by Catalán (2012), the values
of formants F1 and F2 for the Mapudungun vowels are
The area of the sub-palatal space must be presented, values similar to those in our study for F2 in
addressed in a broader sample to identify whether it men and women, but generally with lower values for F1
significantly correlates with the displacement of the than found in our study.
tongue and the values of the formants. In this case, a
larger sub-palatal area was observed in one of the par- This study provides objective and novel data
ticipants, who simultaneously presented greater values about the articulatory analysis of the vowels of Mapudun-
of distance of the tongue to the palate in general. The gun, complementing the information published to date
211Álvarez, G.; Ruiz, M.; Arias, A.; Lezcano, M. F. & Fuentes, R. Description of acoustic and articulatory parameters of vowels in mapudungun speakers. Pilot study . Int. J.
Odontostomat., 14(2):205-212, 2020.
on the acoustic data and contributing a novel methodology 31(2):132-5, 2005.
Echeverría, M. S. Descripción fonológica del mapuche actual. Bol. Filol. Univ.
for their study. However, one of its limitations is the diffi- Chile, 16:13-59, 1964.
culty of accessing participants who speak the language. Franca, M. C. Acoustic comparison of vowel sounds among adult females. J.
To continue this line of enquiry, it would be interesting to Voice, 26(5):671.e9-17, 2012.
Fuentes, R.; Arias, A.; Saravia, D.; Lezcano, M. F. & Dias, F. J. An innovative
study the consonants and the coarticulation of vowels and method to analyse the range of border mandibular movements using 3D
both stressed and unstressed consonants in this language. electromagnetic articulography (AG501) and MATLAB. Biomed. Res.
Future research should assess the articulatory and acoustic (India), 28(9):4239-47, 2017.
Ishida, R.; Palmer, J. B. & Hiiemae, K. M. Hyoid motion during swallowing:
parameters of Mapudungun speakers in the Region of La factors affecting forward and upward displacement. Dysphagia, 17(4):262-
Araucanía and include the participation of other anatomical 72, 2002.
structures in the production of sound in the analysis. Ladefoged, P. Vowels and Consonants?: An Introduction to the Sounds of
Languages. Oxford, Wiley-Blackwell, 2001.
Lee, J.; Littlejohn, M. A. & Simmons, Z. Acoustic and tongue kinematic vowel
space in speakers with and without dysarthria. Int. J. Speech Lang. Pathol.,
19(2):195-204, 2017.
ACKNOWLEDGEMENTS. This research is part of Meenakshi, N.; Yarra, C.; Yamini, B. K. & Ghosh, P. K. Comparison of speech
the project DFP18-0037 and this work was (partially) quality with and without sensors in electromagnetic articulograph AG 501
financed by the Research Office, Universidad de La recording. Singapore, INTERSPEECH. 15th Annual Conference of the
International Speech Communication Association, 2014. pp.935-9.
Frontera. Noiray, A.; Iskarous, K. & Whalen, D. H. Variability in english vowels is com-
parable in articulation and acoustics. Lab. Phonol., 5(2):271-88, 2014.
Rendall, D.; Kollias, S.; Ney, C. & Lloyd, P. Pitch (F0) and formant profiles of
human vowels and vowel-like baboon grunts: the role of vocalizer body size
Álvarez, G.; Ruiz, M.; Arias, A.; Lezcano, M. F. & Fuen-
and voice-acoustic allometry. J. Acoust. Soc. Am., 117(2):944-55, 2005.
tes, R. Descripción de parámetros acusticos y articulatorios de Sadowsky, S.; Painequeo, H.; Salamanca, G. & Avelino, H. Mapudungun. J.
las vocales en hablantes de mapudungun. Estudio piloto. Int. J. Int. Phon. Assoc., 43(1):87-96, 2013.
Odontostomat., 14(2):205-212, 2020. Sánchez Perez, M. A. & Salamanca Gutierrez, G. F. El mapuche hablado en
Lonquimay: Fonemas segmentales, fonotaxis y comparación con otras
RESUMEN: El Mapudungun es un lenguaje utilizado por variedades. Lit. Lingüíst., (31):295-334, 2015.
los mapuches en algunas regiones de Chile y Argentina. El objetivo Soto-Barba, J.; Lara, I. & Salamanca, G. Descripción fonético-acústica de
de este estudio fue describir los fonemas vocálicos respecto a los la sexta vocal en el chedungun hablado en Alto Bío-Bío. Onomazein,
34:229-41, 2016.
parámetros articulatorios (posición de la lengua respecto al paladar
Steele, C. M. & Van Lieshout, P. H. H. M. Use of electromagnetic midsagittal
y apertura mandibular) y los parámetros acústicos (f0, F1, F2 y F3) articulography in the study of swallowing. J. Speech Lang. Hear. Res.,
en hablantes de Mapudungun en la Región de La Araucanía, los 47(2):342-52, 2004.
fonemas vocálicos de Mapudungun son seis, donde los primeros Sun, J.; Yan, N. & Wang, L. Constructing a Three-Dimension Physiological
cinco son similares a los utilizados en español (/a e i o u /), a los Vowel Space of the Mandarin Language using Electromagnetic Articulo-
que se agrega una sexta vocal (/ɨ/) con sus alófonos vocálicos [ɨ] y graphy. Kaohsiung, 2013 Asia-Pacific Signal and Information Processing
[Ә]. Se evaluaron tres hablantes de Mapudungun. Los movimientos Association Annual Summit and Conference, APSIPA, 2013.
de la lengua fueron registrados por Articulografía Electromagnética Toutios, A.; Ouni, S. & Laprie, Y. Estimating the control parameters of an arti-
culatory model from electromagnetic articulograph data. J. Acoust. Soc.
3D y los datos fueron procesados con el software MATLAB y PRA-
Am., 129(5):3245-57, 2011.
AT. Fue posible describir la trayectoria de cada tercio de la lengua Wang, J.; Green, J. R.; Samal, A., & Yunusova, Y. Articulatory distinctiveness
durante la producción de las vocales. Se observó que la sexta of vowels and consonants: a data-driven approach. J. Speech Lang. Hear.
vocal /Ә/ tenía una apertura mínima de la mandíbula durante su Res., 56(5):1539-51, 2013.
pronunciación. Además, se corroboró la característica de /Ә/ como Wang, J.; Samal, A.; Green, J. R. & Carrell, T. D. Vowel Recognition from Arti-
vocal central media no redondeada. En este estudio, la lengua de culatory Position Time-Series Data. Omaha, 3rd International Conference
los hablantes de mapudungun estaba en una posición más posterior on Signal Processing and Communication Systems, 2009.
que la encontrada en otros estudios. Weirich, M. & Simpson, A. Investigating the relationship between average
speaker fundamental frequency and acoustic vowel space size. J. Acoust.
Soc. Am., 134(4):2965-74, 2013.
Palabras Clave: fonética acústica, fonética arti- Zúñiga, F. Mapudungun. El Habla Mapuche. Santiago de Chile, Centro de
cular, Mapudungun, articulografía electromagnética, PRAAT. Estudios Públicos, 2006. Available from: https://www.cepchile.cl/cep/site/
artic/20160304/asocfile/20160304094227/libro_Mapudungun_Fernan-
do-Zuniga.pdf
REFERENCES
Corresponding author
Prof. Dr. Ramón Fuentes Fernández
Álvarez, G.; Dias, F. J.; Lezcano, M. F.; Arias, A. & Fuentes, R. A novel
three-dimensional analysis of tongue movement during water and sali- Research Centre for Dental Sciences CICO
va deglutition: a preliminary study on swallowing patterns. Dysphagia, Dental School
34(3):397-406, 2019. Universidad de La Frontera
Catalán, R. Estudio Fonético Acústico de las Vocales del Mapudungun. Master´s
Av. Francisco Salazar 01145
thesis. Santiago de Chile, Universidad Católica de Chile, 2012.
Chen, W. R.; Chang, Y. C.; & Iskarous, K. Vowel coarticulation: landmark statis- Temuco-CHILE
tics measure vowel aggression. J. Acoust. Soc. Am., 138(2):1221-32, 2015. Received: 25-11-2019
Díaz, A. S.; Cisternas, L. P. & López, B. I. Características acústicas de las vo- Accepted: 06-01-2020
cales del español de Chile producidas por sujetos residentes en la ciudad E-mail: ramon.fuentes@ufrontera.cl
de Santiago. Rev. Chil. Fonoaudiol., 14:92-102, 2015.
Echeverría, M. S. & Contreras, H. Araucanian phonemics. Int. J. Am. Linguist.,
212You can also read
